Powder Blending in Pharmaceutical Production

Content uniformity is a release test, and it is taken at the tablet or the capsule, which is several transfers downstream of the blender. A fine active in a coarser excipient separates at every one of those transfers, and the result arrives as an out of specification result on a batch that was uniform when it left the mixer.

Over blending is the less obvious version of the same problem. A blend of unlike particles is not stable; given enough agitation it will order itself, and a cycle extended to fix a marginal result can produce a worse one. The blending time is therefore a window with a lower and an upper bound, and both have to be established.

Containment runs alongside. The active is frequently potent, so charging, discharging and sampling are all places where the operator meets the compound, and the arrangement that protects them often adds transfers and drops that work against the blend.

Then there is proof. A blend uniformity result is only as good as the sampling, and a thief sample taken from a static bed answers a different question from the one the press will ask.

In pharmaceutical production the blend has a numerical acceptance criterion, and the usual remedy for missing it makes the result worse.

A blend that meets content uniformity at the point of compression, with the transfer route designed to preserve it rather than assessed afterwards.

A blending window, with a lower and an upper time bound, established by experiment and documented with the data behind it.

A sampling plan whose samples represent what the press receives, and a containment arrangement that does not add avoidable transfers.

In plastics compounding the ratio of particle sizes governs: a fine additive blended into coarse resin without a matched carrier separates again on the way to the extruder.

Formulation and equipment are worked on together, because the cheapest fix for segregation is usually in the formulation. Bringing the particle size of the active closer to the excipient, or granulating rather than blending directly, removes the driving force instead of managing it.

Blender selection then follows the material. Tumble blenders suit free flowing systems and are gentle; convective mixers handle cohesive materials and deagglomerate; high shear is for granulation rather than for blending. Blend time is established experimentally, with samples through the cycle so both bounds of the window are visible.

After the blender the design task is restraint. Minimal drop heights, mass flow intermediate containers, short direct transfers to the press and no pneumatic leg if it can be avoided. An in bin blender that becomes the press feed container removes the transfer altogether, which is why it is popular.

Sampling is designed last and deliberately. Samples across the discharge over time, or stratified samples from the press feed, represent what the tablet sees; a thief in a static bed does not, and the difference has been the cause of more investigations than any blender.

• Particle size of active and excipient brought closer, or granulation used, before equipment is blamed.
• Blender type matched to the material: tumble for free flowing, convective for cohesive.
• Blend window established by sampling through the cycle, with both bounds documented.
• Transfers after the blender minimised, with mass flow containers and no unnecessary drops.
• In bin blending considered to remove the post blend transfer entirely.
• Sampling plan designed to represent the press feed, not the static bed.

• Content uniformity requirement and the sampling plan that will have to prove it.
• Particle size and density difference between active and excipients, which is the driving force for segregation.
• Transfer route from blender to press, counted in drops and metres.
• Containment band of the active, which decides charging, discharging and sampling arrangements.
• Cohesiveness of the formulation, which decides between tumble and convective mixing.
• Cleaning validation for the blender and every transfer item in the route.